Throughout TB Alliance’s history, it has expanded its pursuit and advocacy for various types of TB research, as it identifies opportunities to achieve meaningful progress an impact. In one example of this, TB Alliance recently relaunched MARK-TB, an initiative providing researchers with access to a high-quality tuberculosis (TB) biorepository to facilitate advances in TB biomarker research.
The field of TB biomarker research holds great potential to impact TB control on all vectors. The discovery, validation, and development of products to measure TB biomarkers could unlock paths to vaccine development, advances in TB diagnosis, and acceleration of TB drug development by reducing the length and cost of trials needed to approve new therapies.
To offer insight into the basics and potential of TB biomarkers, we interviewed Riccardo Alagna, a Research Fellow and public health specialist in the Emerging Bacterial Pathogens Unit at San Raffaele Scientific Institute in Milan. The following conversation has been condensed and edited for clarity.
What is a biomarker? A biomarker is any defined characteristic that can be measured and used as an indicator of biological processes. It can be something that is normally happening in your body, or that happens during the development of disease, or in response to a particular medicine. Usually biomarkers are biological molecules you can find in the blood, in body fluids, in sputum (as in the case of tuberculosis), or in tissues. In general, a biomarker is something that you can use to measure a process in an objective way, so you know that a particular biomarker is correlated with a specific result.
How are biomarkers used in TB? TB biomarkers have several applications. However, their applications are still limited due to the lack of large trials. Biomarkers could be used to understand whether a given host following vaccination has become immune, or to detect active TB, or to understand whether a person has an increased risk of developing active TB, or, ultimately, to understand whether a patient is responding effectively to anti-tuberculous treatment.
At the moment, biomarkers are mainly used to identify an active disease. Some examples are technologies based on DNA detection of Mycobaterium Tuberculosis in sputum, or non-DNA biomarkers such as the lipoarabinomannan antigen detected in urine, both used to diagnose TB.
A second application of biomarkers is, for clinical trial, such as the culture-based biomarkers - Early Bactericidal Activity, sputum culture conversions, and so forth. However, these biomarkers have limited capacity to predict regimen success.
How could finding new biomarkers help improve TB control? If you have biomarkers, you have potential to improve all the three areas of disease control: prevention, diagnosis and treatment.
In prevention, the most powerful weapon to end a pandemic is a vaccine. Biomarkers would be valuable in pursuing vaccines because they could be helpful in indicating immunity to developing TB disease and shortening the development time. In diagnosis, a biomarker to identify the disease as early as possible and whether you have high risk of developing active TB disease, would be a cornerstone. And, in treatment, if you find biomarkers that can speed up development of new medicines, you’ll find new and better cures faster.
How would new biomarkers accelerate the development of new treatments? A biomarker able to predict the clinical outcome could be used to evaluate new regimens. All culture-based biomarkers used along the different phases of TB drug development are imperfect surrogate endpoints for the evaluation of new treatments in a clinical trial. A new validated biomarker working as a surrogate endpoint could speed up the clinical trial because you can detect outcomes.
It could also streamline clinical trials moving toward personalized medicines by stratifying patients with particular sub-types (i.e. hard-to-threat phenotype or low bacterial burden or absence of lung cavities). You can also improve trial monitoring and stop unhelpful interventions early.
Last but not least, it can speed up the regulatory pathway for drug or regimen licensing. However, to do this, you need biomarkers that can act as surrogate endpoints and reliably predict the differences in clinical outcome between two different treatments.
How difficult is it to identify and validate a biomarker for TB? Once you find a biomarker, you need to validate it. This is very challenging, as there are often inconsistencies between studies of the identified biomarker; data quality is an issue. However, once you validate a biomarker you then have to translate that to a test that can be used in settings with limited infrastructures. In TB, that means it needs to be, if possible, portable and usable at the point of care, user-friendly, and with high reliability to identify or measure that biomarker. So, challenges span from basic research, to translational research, to clinical application.
What are the main challenges to tuberculosis biomarker research? The issue is multi-faceted. The main barrier is historically insufficient funding – annual TB research investments never approached half of the US$2 billion needed. A second factor is the dynamic of pathology; the spectrum of TB pathology is complex. The differences between containment and progression to active disease, and the immunological response to TB are not completely clear. Different biomarkers apply to different states of the disease, but a person may actually move back and forth between those different states. A third factor is that biomarker research is a bit fragmented. We need collaboration between basic research scientists, clinical trial investigators, the pharmaceutical industry or manufacturers, and the end user.
How would better collaboration improve the field? When you study biomarkers and you want to validate them, you need to be sure that your sample collection processes are standardized, and you have captured all the clinical data. In biomarkers, it is like working as pioneers in the Wild West of bioanalysis. But once something is found and published, we need to replicate the results. If processes are not standardized or clinical data capture is poor, you cannot reproduce results and cannot validate it.
What are specific challenges to discovering and validating biomarkers to predict treatment outcomes? I was mentioning that biomarker research is like working in the Wild West of bioanalysis. Most of biomarker research is characterized by small sample numbers, variation between study cohorts and geographical location of study, poor quality control of laboratory techniques and heterogenous datasets. This means only a small proportion of biomarkers research is assayed in a clinically relevant manner.
When you look for treatment outcome, you need to look at long term results. So, you require specimens with all the clinical data and microbiological information from patients. Then, you need to have adequate follow ups. You need to be working within well-structured studies that provide this information reliably. This means you need a truly long-term collaborative effort and adequate resources.
How does a biobank, like the one now available to researchers through MARK-TB, advance biomarker research? A well-established and curated biobank is the solution to address challenges in biomarkers studies. It allows researchers to have access to precious material and high-quality data. It regulates the type, quality, and method of data collected and acts as a model to the research community, showing them how to implement the same research. Years back, I think biobanks were just considered a storage of samples, but now the biobank is much more. It’s an entity that can overcome some of the shortfalls in biomarker research I’ve mentioned – so they establish protocols, quality control, and address the effects of preanalytical and storage variation on a broad range of sample types. We need to raise awareness that biobanking is key to biomarker research.